Re: Help, my developers are killing me with varchar2(4000)

sybrandb_at_hccnet.nl wrote:
> On Mon, 11 Aug 2008 13:21:42 +0000 (UTC), Wolfram Roesler <wr_at_spam.la>
> wrote:
>
> > Neither has the question been asked why Oracle doesn't support
> >a random length string datatype.
>
> That question doesn't need to be asked.
> First of all Oracle already has CLOBs for this purpose, since 8.0, so
> in the previous milennium.

Using Perl's DBD::Oracle, CLOBs have several undesirable side effects if what one wants is a varchar2 but without the arbitrary limit. I don't know if other drivers do a better job of circumventing these problems or not.

> Secondly, a random length string datatype would cause heavy row
> chaining, so kill performance.

If 99.99% of the data has short strings and 0.01% needs row chaining, the performance impact will be negligible, while the benefits of not having 0.01% of your data trigger the collapse of your application are great.

The A, the I, and the D of ACID all "kill" performance, too. Yet they are worth it.

Xho

-- 
-------------------- http://NewsReader.Com/ --------------------
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby marked
advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate
this fact.